JPH04501157A - Method and apparatus for identifying misfires - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed description of the invention] Method and apparatus for identifying misfires The present invention relates to a method and apparatus for identifying misfires in internal combustion engines.

Conventional technology Conventionally, misfires are detected in two different ways, specifically in each series of an internal combustion engine. It is detected by measuring the pressure in the conductor or by measuring the light. different types The measured values of the cylinders are fed to a common control unit and processed there. The equipment becomes complicated. Because of this, it has not been used in mass-produced cars until now.

Particularly in vehicles equipped with catalysts, it is preferable to be able to continuously monitor misfires. this is This is because due to a misfire, unburned fuel enters the catalyst and ignites there. child This often burns out the catalyst and, in particularly severe cases, can cause a vehicle fire. Resulting in.

The object of the invention is to provide a method for identifying misfires that is suitable for mass production.

A further object of the invention is to provide a device of simple construction for carrying out this method. Ru.

Advantages of invention The method according to the invention is characterized by the features of claim 1 and the apparatus is characterized by the features of claim 10. given by the characteristics of the term. Preferred embodiments and configurations of this method are included in the claims. It is described in Sections 2 to 9.

Although the method of the present invention is a pressure measurement method, it does not measure the combustion pressure of each cylinder. The pressure in the exhaust gas path is detected. This provides two advantages. one of them is that all cylinders can be monitored with a single pressure sensor, other advantages are: It is possible to detect not only ignition defects but also exhaust gas path defects. usually carried out Whether or not a method can achieve both advantages depends on the implementation used. Exclusion If the gas pressure is sensed by a separate sensor located in each manifold, the configuration The advantage of simplicity is lost. A specific method that reduces pressure This may have been caused by an ignition defect or an opening in the wrong part of the exhaust gas line. If you do not consider whether these defects are caused by The advantage of being separate is lost.

In order to detect defects, the method according to the invention measures the exhaust gas pressure after each ignition stroke. compared with the waiting value. If the difference between both values is greater than a predetermined difference, an event signal is output. be done. The frequency of this signal is monitored and if a frequency threshold is exceeded, a signal indicating a defect is generated. Output. The defect signal is used, for example, to light up a warning lamp. but , preferably in embodiments of the method, it is possible to accurately detect what caused the defective signal. If the detection is successful, appropriate defect handling measures are taken.

The frequency of the event signal that causes the defect signal to be output is chosen purposefully. In extreme cases, This is a case where a defect signal is output immediately when an event signal occurs, and therefore the comparison is The frequency of occurrence is a -stroke event signal. However, usually a predetermined number, e.g. 10 If a predetermined number of event signals, for example 10, are exceeded within the 00 process, it is determined to be a defect. is preferred.

Ignition defects or exhaust gas path defects are investigated in particular according to two principles: . One is that if a predetermined frequency threshold, which is a relatively small value, is exceeded, the individual It is assumed that each cylinder has an ignition defect. In contrast, the event signal If the frequency is very large, this is caused by a defect in the entire ignition system. or the exhaust gas path is open (i.e., the exhaust pipe is open at a point other than the end). (released). In this case, the exhaust gas pressure in case of misfire The second principle is used, in which the temporal value of the force is different from when the exhaust gas path is open. It will be done. Therefore, the actual value is compared with the predetermined value and the event is A defect signal belonging to is output.

It is not enough to simply check whether the measured value of the exhaust gas pressure is below the expected value by a predetermined difference. When determining whether a measured value exceeds the expected value by more than some other predetermined threshold, In this case, it is possible to check whether the exhaust gas path is clogged.

The device according to the invention has means for carrying out the method described above. This can be achieved using a built-in processor. Wear.

drawing The present invention will be described in detail below based on embodiments shown in the drawings.

FIG. 1 is a block diagram of the device according to the invention in an internal combustion engine.

FIG. 2 is a flow diagram illustrating the flow of a preferred method for identifying misfires.

Description of examples The misfire identification device shown in the block diagram of FIG. An internal combustion engine 10, shown schematically without distinction, is monitored. Intake path 11 and exhaust gas path 12 are also each shown as just a simple tube without branching into a cylinder.

A throttle valve 13 is arranged in the intake passage 11, and its position is determined by an angle sensor of arbitrary configuration. Detected by The position of the throttle valve indicates the load under which the internal combustion engine lO is driven.

The load signal can also be obtained in other known ways. There is a pressure signal p in the exhaust gas path 12. A pressure sensor 15 that outputs M(p) is arranged. A device that identifies misfires As signals to be further processed, the crank angle δ and rotation speed n are shown in FIG. It is shown.

In each cylinder of an internal combustion engine, the corresponding exhaust valve device opens at a predetermined crank angle δ. Ru. Due to the cam geometry, this opening is initially slow and then accelerated. that The amount of exhaust gas discharged into the exhaust gas path is slow at first, then increases rapidly, and then After that it decreases. Normally, the exhaust gas flow at the exhaust port is already reduced before the exhaust valve device is fully opened. The maximum value has been reached. When the exhaust gas flow at the exhaust valve device changes over time, Associated pressure changes over time.

Pressure and exhaust gas propagate at different speeds. In other words, pressure propagates at the speed of sound, while The emitted exhaust gas travels at a predetermined flow rate. Pressure propagation is important because: pressure The force signal includes the exhaust valve device and the pressure sensor 15 from the exhaust valve device to the pressure sensor 15. The time required corresponds to the product of the distance between the sensors and the speed of sound. Various rotation speed n and load If the value of is taken, compare at which crank angle δ the pressure at the exhaust valve device is maximum. accurately known. Furthermore, pressure signals are transmitted from each exhaust valve device to a pressure sensor. Since we know exactly how long it takes to reach each crank angle mentioned above, At what point in time is the maximum pressure developed in a given cylinder detected by the pressure sensor 15? You will know exactly what you will reach. Ideally, the pressure sensor reading should be exactly at this point. I can't stand it.

To make this possible, FIG. 1 shows measurement angle map values 16, time signal generator 1 7 and a sample and hold circuit 18 are shown. The measurement angle map value is 16. Actual values of crank angle δ, rotational speed n, and load are constantly input. Rotation speed n and load From map value 16, the exhaust gas pressure at the exhaust valve device is maximized according to the The angle is read out. The actual crank angle δ is the crank angle at which the exhaust gas pressure is maximum. When it matches δM, a single signal is output to the time signal generator 17. time signal The generator allows the pressure to be measured between the exhaust valve device of the cylinder whose pressure is to be measured and the pressure sensor. The corresponding time is set. After the set time has elapsed, this is the measurement point. This means that it has become tM. Corresponding trigger signal from time signal generator 17 In response, the sample hold circuit 18 detects the actual exhaust gas pressure value pM (tM). put out

According to the respective rotational speed n and load value, the pressure sensor is It is known what value the sensor 15 will measure. This expected exhaust gas The exhaust gas pressure value pK is read out from the exhaust gas pressure value 19, and the comparator stage 20 reads this value. The measured exhaust gas pressure pM (tM) is subtracted from the value of . The difference is important for defect identification and missing parts. It is input to the error processing means 21. This means that the difference signal is analyzed and the flow described below is to determine if there is a defect in the fuel supply or ignition. Results of this survey Various measures will be taken accordingly.

Note that each block shown in Figure 1 is a programmed microprocessor. This can be realized by

The process flow outlined based on the device in Figure 1 is detailed according to the flowchart in Figure 2. Explain.

When the method starts, various values are initialized between marks A and 8. Ste The exhaust gas pressure pM is detected in step sl at measurement time tM. The measurement time tM is How this is determined has already been explained in FIG. step s The expected value pK of exhaust gas pressure, which has already been explained in detail in 2, is the characteristic map value. 19. In step s3, the measured value pM of the exhaust gas pressure is set to a predetermined difference. 690 or more, whether the expected value of exhaust gas pressure pK has been exceeded, that is, the exhaust gas passage is clogged. It is determined whether the If the result is YES, the pressure increase signal is Corresponds to what is being output. In step s4, this pressure increase signal is You can check how often it appears within the equation. This is the measurement frequency HPGM. Ru. This frequency is compared with a threshold frequency HPGS in step s5. threshold frequency When the degree is reached or exceeded, a warning lamp is lit in step s6, A diagnostic signal indicating that the exhaust gas path is blocked is stored.

On the other hand, if the frequency is below the threshold frequency, the process goes to step s7. child This step is performed during the determination of whether the exhaust gas passage is clogged in step s3. This is also the next step after step s3 when the result is negative. In step s7 If the measured value of exhaust gas pressure is lower than the expected value of exhaust gas pressure by more than a predetermined difference Δpu, You can check to see if it is spinning. If this judgment is denied, mark B Return to

On the other hand, if the judgment in step s7 is affirmative, the measured exhaust gas This indicates that the pressure is lower than expected. This is because the ignition Either this is not done or the exhaust gas path is open. Which one is it? Another method is to check whether the flag SP is set in step S8. It is started by. This flag is initialized between marks A and 8, so Initially not set. If it is not set, go to step s9. , where the frequency HPKM is measured. This measured frequency HP in step slO KM is compared to a threshold frequency HPKS.

In contrast, it is detected in step S8 that the flag SP is set. If so, the frequency HPKSPM is measured, and in the following step s12 the threshold frequency H Compared to PKSPS.

Frequency HPGM of step s4, frequency HPKM of step S9 and step sl The frequency HPKSPM of l, and the threshold frequency HPGS of step s5, step Threshold frequency HPKS of slo and threshold frequency HPKS of step s12 Please note the following regarding PS.

A plurality of event signals are measured, each within a predetermined number of strokes. When the event signal is measured , becomes positive (pressure increase signal) during step s3 or during step s7. is positive (pressure decrease signal). If there is a defect in the exhaust gas line, it may be clogged. Whether due to leakage or undesired opening, the event signal This occurs almost every step. In contrast, only individual cylinders lack ignition. In some cases, event signals appear less frequently. How does this difference affect the frequency? and how to set the corresponding thresholds. Ru. The lack of ignition in individual cylinders is an unsteady condition even if there is no defect in the fire system. This is because misfires often occur in the transition region.

In this example, the frequency HPKM is the amount of pressure reduction signal within 1000 strokes. This can be determined by checking whether this has occurred. This is 100 strokes with a counter. This is done by determining how much pressure reduction signal is generated within the range. 10 This counter is reset every 0 stroke, and the count value l/lO is 9/lO will be added to the total. The new sum of 9/lO is used during the next addition. this In this way, the results of the latest 1000 runs can be used at all times. Do it like this The threshold frequency HPKM has a value of lO at intermediate loads and intermediate speeds. Compared to HPKS.

Other frequencies HP used to detect defects not based on individual cylinder misfires The values of GM and HPKSPM are determined in only 100 strokes in the preferred embodiment.

The threshold frequency HPGS when the pressure is unexpectedly high is 50 confidences in 100 strokes. This is the number. The threshold frequency HPKSPS is defined as the ratio of the pressure signal to the stroke signal over the whole series. It is chosen to be somewhat larger than the cylinder to cylinder ratio. 4 cylinders In the case of an engine, this ratio is 1:4, so the threshold frequency is 100 strokes. Select 30 pressure reduction signals.

As mentioned above, in step slO, the frequency HPKM reaches the threshold frequency HPKS. You will be judged on whether you have done it or exceeded it. If not, processing returns to B. child If this is the case, the flag SP mentioned above is set and the subsequent processing similarly returns to B. When step S8 is reached again, flag SP is set this time. Therefore, the process branches to step sll. In step s12, the frequency Check whether HPKSPM has reached or exceeded the threshold frequency HPKSPS. It will be done.

The reason for the processing sequence described above is as follows.

Reaching or exceeding threshold frequency HPKS in step slo is detected, the defect is distributed evenly over the previous 100 runs each examined. It is unclear whether the defects appeared more frequently in the past, or whether the defects were more frequent in recent processes. do not have. Assuming that each cylinder temporarily lacks ignition, the 1000 cylinders examined last time During the stroke, e.g. 11 pressure reduction signals appear gradually. On the other hand, during the 989th journey If no pressure decrease signal appears, 11 pressure decrease signals appear in succession. this is This indicates that the ignition of individual cylinders does not have to be defective. this place A distinction is made between steps sll and s12.

In step s12, the frequency HPKSPM is smaller than L/threshold frequency HPKSPS In some cases, this indicates that only the ignition of an individual cylinder is defective. Continued In step s13, a warning lamp is turned on and a corresponding diagnostic signal is stored. Change Defect countermeasures will be taken.

In the preferred embodiment, the defect frequency is only common to all cylinders. rather than the measured value of each cylinder. This is not a problem . This is because when the pressure sensor 15 is activated, each one of the cylinders Each pressure sensor operates when the amplitude of the pressure wave from the cylinder passes. It is. The main defect countermeasure in step s13 is to remove the defective cylinder. The purpose is to cut off the injection. As a result, unburned fuel is no longer supplied to the catalyst. Ru. In step s14, after the waiting time has elapsed, the defective cylinder is also Injection starts again. However, in that case, the ignition energy for this cylinder is Increased. Processing returns to mark A.

If the defect is not resolved by increasing the ignition dwell time, the defect Long-term cut off of injection to certain cylinders. On the other hand, there are no defects If the standby time has elapsed, the system will switch back to normal operation. defect again If this occurs, increase the ignition dwell time in the long term. In either case, a warning Keep the lamp on all the time. However, if the defect no longer occurs, the warning Turn off the lamp and clear the diagnostic signal.

Note that countermeasures for defects related to ignition dwell time and injection adjustment may be taken individually. or only in the reverse order.

In step s12, it is determined that the frequency HPKSPM is greater than or equal to the threshold frequency HPKSPS. If you find that A power reduction signal often indicates a total ignition system failure or an open exhaust gas line. It indicates that something is happening. Ignition defect Exhaust gas path defect In order to distinguish from Measure. Based on FIG. 1, the middle measurement is performed at the time tM explained in step s1. be exposed. The first measurement was taken a predetermined time earlier, and the second measurement was This is done after a predetermined period of time.

In step s16, the ignition is operating normally but the exhaust gas path is open. The expected pressure value DKK that will occur in this case is calculated. This expected pressure value is already shown in Figure 1. In the same way as explained for the expected value of each exhaust gas pressure in step s2 based on It can be read from the gender map value. However, in the preferred embodiment, It can also be done. In other words, the exhaust gas pressure determined during the middle measurement in step s15 The measured force value is used as the initial value. This value predicts the prediction at the first and third measurement points. The expected value is calculated. If there is a defect in the exhaust gas line, the above-mentioned for the maximum pressure The pressure will be much lower than if the ignition was defective.

In step s17 the measured pressure value is compared with the expected pressure value. This example Now, this comparison is based on the pressure values DVM obtained by the first and third measurements and the corresponding periods. This is done by comparing the wait value DVK. Predetermined limit value ±ΔD for both values If the comparison values agree with each other within V, this indicates a defect in the ignition. are doing. In this case, the warning lamp is turned on in step s18 and the diagnostic signal is pay. If there is a significant lack of ignition, the vehicle will normally be unable to function. Therefore, no countermeasures can be taken against defects.

On the other hand, if the answer is NO during the determination in step s17, the step In step s19, a warning lamp is lit in the same way, but the exhaust gas path is unfavorable. A diagnostic signal is output indicating that the device is open at the location. The only meaningful measure is to The solution is to close the incorrect opening again, and this measure can be taken automatically. Since there is no defect, no countermeasures can be taken.

Note that under certain operating conditions, a standing wave is generated in the exhaust gas path, and this standing wave is caused by the cylinder. May continue even in the absence of ignition. In this case detecting the lack of ignition I can't. Therefore, the signal from the pressure sensor may not accurately indicate the investigated defect. It is preferable to stop frequency measurement in operating conditions where it is known that That time The counting of strokes and the defective signals that occur are not counted. Preferably , the operating state processed in this way is stored in the characteristic map value. map value is satisfied Stop frequency measurements whenever the frequency is exceeded.

It is also possible to calculate the value obtained using mathematical operations, but in this case, the calculation time is It will take a while.

The sensor for pressure measurement is preferably a piezoelectric sensor, which has thus far been used in cylinders. This is a sensor that was used to measure pressure.

-ni-C' Copy and translation of amendment) Submission (Article 184-8 of the Patent Law) April 4, 1991

Claims (1)

[Claims] 1) Obtain the measured value of exhaust gas pressure for each stroke, and calculate the value for each stroke from the operating parameter values. Find the expected value of exhaust gas pressure for Compare the measured value of the exhaust gas pressure and the corresponding expected value of the exhaust gas pressure. , outputs an event signal when the two values differ from each other by more than a predetermined difference; determining the frequency of the event signal; An internal combustion engine that outputs a defect signal when the frequency exceeds a threshold value. How to identify a misfire. 2) When the measured value becomes smaller than the expected value by more than a predetermined difference, it is a pressure decrease signal. A method according to claim 1, characterized in that an event signal is output. 3) According to claim 2, wherein the defect signal is a misfire defect signal. the method of. 4) when the frequency is below a threshold, outputting a misfire defect signal; If the frequency exceeds the threshold, a fault signal of an ignition fault or a fault signal of an exhaust gas path is issued. 3. The method according to claim 2, further comprising outputting a code. 5) Find the actual exhaust gas pressure value for at least one stroke and expect each value found. Compare it with the exhaust gas pressure value, and as a result of the comparison, the measured value indicates that the ignition is normal but the exhaust gas path is When the value approximately corresponds to the expected value when the exhaust gas path is open, an exhaust gas path open signal is output. , On the other hand, as a result of the comparison, the measured values indicate that there is a defect in the ignition and that the exhaust gas path is operating normally. output an ignition defect signal when the case approximately corresponds to the expected value; Claims characterized in that a defect signal of an ignition defect or a defect signal of an exhaust gas path is obtained. The method described in Scope No. 4. 6) When the measured value of exhaust gas pressure exceeds the expected value by a predetermined difference, a pressure increase signal is sent. output, calculate the frequency of this signal, and when the calculated frequency exceeds the threshold, the exhaust gas path Any one of claims 1 to 5, characterized in that a clogging signal is output. or the method described in paragraph 1. 7) When used in an internal combustion engine that performs injection sequentially, the above method may be used for all injections. In addition to performing them together during the firing stroke, it is also recommended to perform them individually for each cylinder's ignition stroke. A method according to any one of claims 1 to 6, characterized in that: 8) When a defective misfire signal is obtained by the individual cylinder identification method, the defective The method according to claim 7, characterized in that the injection to the cylinder is cut off. Law. 9) When a defective signal of a misfire or a defective signal of an ignition defect is obtained, reduce the ignition energy. According to any one of claims 1 to 8, characterized in that Method described. 10) A sensor (15) that measures the exhaust gas pressure and the measured value of the sensor during each stroke means (16, 17, 18) for detecting; Means for determining the expected value of exhaust gas pressure for each stroke from the value of the operating parameter (19 )and, Compare each measured value of exhaust gas pressure with the corresponding expected value, and ensure that both values are correlated with each other by more than a predetermined difference. means (20, 21) for outputting an event signal when the difference occurs; means (21) for determining the frequency of the event signal; and means (21) for outputting a fault signal. Separate equipment.